남북사회통합연구원

The Collapse of
North Korea
Bruce W. Bennett and
Jennifer Lind
Military Missions and Requirements

Many signs suggest
that Kim Jong-il’s regime in North Korea is entering a difªcult stage in which
its future may be in doubt. Although the historical record shows, and many
scholars have noted, that authoritarian regimes can repress their populations
and retain power for decades,1 the Kim regime is embarking on the most
difªcult challenge that such regimes face: succession.2 The last time power
changed hands in Pyongyang, Kim Il-sung spent ªfteen years preparing for
the transfer, carefully consolidating support for his son Kim Jong-il. By contrast,
Kim Jong-il, who suffered a stroke in 2008, has only recently anointed his
inexperienced, twenty-seven-year-old third son, Kim Jong-un, as his heir.3 Kim
Jong-il’s sudden death or incapacitation could trigger a power struggle and
government collapse in North Korea.4 As previous revolutions in the Middle
East and Eastern Europe demonstrate, the transition from apparent stability to
collapse can be swift.
A government collapse in North Korea could unleash a series of catastrophes
on the peninsula with potentially far-reaching regional and global effects.
Collapse would likely trigger a humanitarian crisis. Many of North Korea’s
24 million inhabitants are already severely malnourished; if governmentprovided
food and health services were to cease, the population would rapidly
face the prospect of starvation. Food shortages and the possibility of civil war
Bruce W. Bennett is Senior Defense Analyst at the RAND Corporation. Jennifer Lind is Assistant Professor
of Government at Dartmouth College.
The authors wish to thank the following individuals for their help: Kevin Benson, Victor Cha, Robert
Collins, James Dobbins, Bruce Klingner, Col. David Maxwell, John Park, Jonathan Pollack,
Barry Posen, Scott Snyder, and the anonymous reviewers. They thank Daryl Press and his students
Andrew Son and Roger Zhu. They also thank Kathryn Lindquist for research assistance. Finally,
the authors are also grateful for feedback from audiences at the Naval War College, the Security
Studies Program at the Massachusetts Institute of Technology, and the Center for Strategic and International
Studies.
1. Daniel L. Byman and Jennifer Lind, “Pyongyang’s Survival Strategy: Tools of Authoritarian
Control in North Korea,” International Security, Vol. 35, No. 1 (Summer 2010), pp. 44–74; Stephan
Haggard and Marcus Noland, Famine in North Korea: Markets, Aid, and Reform (New York: Columbia
University Press, 2007), pp. 221–228; Marcus Noland, “Why North Korea Will Muddle
Through,” Foreign Affairs, Vol. 76, No. 4 (July/August 1997), pp. 105–118; and Mark McDonald,
“Hardships Fail to Loosen Regime’s Grip in N. Korea,” New York Times, February 24, 2011.
2. Byman and Lind, “Pyongyang’s Survival Strategy,” pp. 71–72.
3. Mark McDonald, “Son of North Korean Leader Is Said to Be Given No. 2 Post,” New York Times,
February 16, 2011.
4. For speculation about divisions in the regime, see Jean H. Lee, “Cheonan Attack May Be Tied to
North Korean Succession,” Christian Science Monitor, May 27, 2010; and Martin Fackler, “Test Delivers
a Message for Domestic Audience,” New York Times, May 26, 2009.
The Collapse of North Korea
The Collapse of
North Korea
Bruce W. Bennett and
Jennifer Lind
Military Missions and Requirements
International Security, Vol. 36, No. 2 (Fall 2011), pp. 84–119
© 2011 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.
84
would trigger a massive outºow of refugees, as desperate North Koreans
searched for food and safety across international borders. North Korea’s weapons
of mass destruction (WMD) could ªnd their way out of the country and
onto the global black market.
If other countries wanted to intervene to mitigate such instability, they
would need to perform complex military operations. The provision of humanitarian
relief could not be delegated to international relief organizations. Because
North Korea has some 1.2 million active-duty military personnel and
7.7 million reservists,5 outside military intervention would likely be necessary
to provide security for such operations.
The consequences of a poorly planned response to a government collapse in
North Korea are potentially calamitous. Rapid cooperation would be essential
because many response missions are time-sensitive—for example, the longer
it takes to organize humanitarian efforts, the higher the number of North
Koreans who might perish or decide to leave their homes; in addition, the
longer North Korean WMD are left unsecured, the larger the risk that they will
disappear across international borders. Perhaps the greatest danger is that
countries will send their militaries in without coordination to stabilize the area
or to secure the WMD. The specter of Chinese forces racing south while U.S.
and South Korean troops race north is terrifying given the experience of the
Korean War, a climate of suspicion among the three countries,6 and the risk of
escalation to the nuclear level.7
Some countries have begun planning for North Korea’s collapse. The United
States and South Korea have negotiated an operational plan for joint military
responses to this and other emergencies that could arise in North Korea.8
South Korea’s president, Lee Myung-bak, has proposed a taxation plan to prepare
for the monumental ªnancial burden of Korean uniªcation.9 China’s
People’s Liberation Army (PLA) reportedly has developed contingency plans
for humanitarian, peacekeeping, and counter-WMD-related missions in North
The Collapse of North Korea 85
5. Ministry of National Defense, Defense White Paper, 2008 (Seoul: Republic of Korea, March 23,
2008), p. 316.
6. On South Korean fears of Chinese territorial revisionism, see Peter Hays Gries, “The Koguryo
Controversy, National Identity, and Sino-Korean Relations Today,” East Asia, Vol. 22, No. 4 (Winter
2005), pp. 3–17.
7. Jonathan D. Pollack warns of an “international crisis, caused more by inadvertence than by design.”
Pollack, “The Korean Peninsula in U.S. Strategy: Policy Issues for the Next President,” in
Ashley J. Tellis, Mercy Kuo, and Andrew Marble, eds., Strategic Asia 2008–09: Challenges and
Choices (Seattle, Wash.: National Bureau of Asian Research, 2009). On the risk of crisis escalation
between China and the United States, see Keir A. Lieber and Daryl G. Press, “The NukesWe Need:
Preserving the American Deterrent,” Foreign Affairs, Vol. 88, No. 6 (November/December 2009),
pp. 39–51; and “Calling Kim Jong Il’s Bluff,” Economist, April 23, 2009, p. 50.
8. “Seoul Overhauls N. Korea Contingency Plan,” Chosun Ilbo, January 14, 2010.
9. Choe Sang-hun, “South Korean Leader Proposes a Tax to Finance Reuniªcation,” New York
Times, August 15, 2010.
Korea.10 Thus far, however, multilateral planning that involves China has been
stymied. Long aligned with North Korea, China has been reluctant to provoke
the Kim regime by coordinating plans for its demise with its enemies. In addition,
the Chinese worry that open discussion of a North Korean collapse could
increase the probability that it occurs.11 A failure to engage in combined planning,
however, could be catastrophic because of the risks of misperception and
crisis escalation.
The purpose of this article is twofold. First, we seek to bring into the public
debate a discussion of the scale of the problems that the collapse of North
Korea’s government could create, and the potential for dire consequences,
both humanitarian and strategic, if stability efforts were delayed or failed altogether.
We describe the military missions that might be necessary to stabilize
North Korea and estimate the force requirements for those missions.12 In these
estimates, we put aside the question of whether South Korean, U.S., or other
troops would conduct the operations. Throughout the analysis, however, we
discuss the interests and potential involvement of various countries.
Second and more broadly, this analysis sheds light on international intervention
in collapsing states. Each case is of course unique, but this article provides
a framework for thinking about this kind of problem—a problem that foreign
policy planners envision as increasingly salient in perhaps Colombia, Iran,
Pakistan, and even Mexico.
Based on optimistic assumptions about how a collapse might occur, we estimate
that 260,000–400,000 ground force personnel would be required to stabilize
North Korea. This means that even in the relatively benign scenario that
we describe, the requirements for stabilizing a collapsed North Korea would
outpace the combined U.S. troop commitments to Iraq and Afghanistan.13
Managing a more demanding Korean collapse scenario would push these requirements
higher or lengthen the duration of the operation, or possibly both.
International Security 36:2 86
10. Bonnie Glaser, Scott Snyder, and John S. Park, “Keeping an Eye on an Unruly Neighbor: Chinese
Views of Economic Reform and Stability in North Korea,” Working Paper (Washington, D.C.:
Center for Strategic and International Studies and U.S. Institute for Peace, January 3, 2008), p. 19;
Drew Thompson, “Border Burdens: China’s Response to the Myanmar Refugee Crisis,” China Security,
Vol. 5, No. 3 (2009), pp. 13–20; and Park Changhee, “North Korean Contingency and Prospects
of China’s Military Intervention,” Working Paper, No. 5 (Seoul: Ilmin International Relations
Institute, Korea University, October 2010).
11. On Pyongyang’s fury at U.S. and South Korean plans for collapse, see Lee Tae-hoon, “NK Regards
OPLAN 5029 as Declaration of Warfare,” Korea Times, November 8, 2009.
12. We conªne our analysis to triage—the most pressing challenges that could arise in the short
term during transition. Challenges that Korea will have to face in the longer term include infrastructure
repair and massive economic investment in the north, institutional and economic reform,
jobs creation, de-indoctrination, education reform, and transitional justice. The country will have
to make decisions about its nuclear posture and its future alliance policy.
13. As of May 2010, there were 94,000 U.S. military forces in Afghanistan and 92,000 in Iraq. Associated
Press, “Number of U.S. Forces in Afghanistan Exceeds Total in Iraq for the First Time,”
Washington Post, May 25, 2010.
This analysis yields several policy implications. First, the magnitude of these
potential requirements (which would probably fall most heavily on South
Korea) should inform contemporary defense planning in Seoul. The demands
that these missions would place on South Korea’s military are at odds with its
planned drawdowns in ground forces and with the current structure of its military
reserves.14 As Seoul contemplates defense reforms, it should do so with
an understanding of the potentially staggering requirements for stabilizing
North Korea.
Second, these possible force requirements underscore the need for advance
planning with China. Seoul and Washington should discuss with each other
and with Beijing the prospect of Chinese participation in missions to stabilize
North Korea after a government collapse. If Seoul and Washington oppose
Chinese involvement, then they should be prepared to conduct these missions
in ways that obviate the need for Chinese intervention. Ultimately, Seoul and
Washington may have no control over a Chinese decision to send the PLA into
a collapsed North Korea. The prospect of unilateral Chinese military action,
and the dangers associated with uncoordinated stabilization efforts, suggest
the importance of advance and combined planning. Although Beijing has so
far resisted such discussions, South Koreans and Americans should at a minimum
expand their track II talks with the Chinese, and they should continue to
press Beijing for ofªcial coordination.
Our analysis proceeds in four sections. We ªrst discuss the potential for stasis
and change in North Korea and describe our assumptions regarding a collapse
of North Korea’s government. The second section details the various
problems that such a collapse could unleash, lays out the military missions
that countries might choose to perform to mitigate those problems, and estimates
the requirements for those missions. In the third section, we examine the
interests of countries that may choose to contribute to stabilization missions.
The fourth section addresses critiques of our argument and outlines several
policy recommendations.
Scenarios and Assumptions
Four different futures for North Korea’s leadership are possible; three of them
would not produce the kind of instability that could prompt international
intervention. First, the Kim regime could “muddle through” for years to
The Collapse of North Korea 87
14. The South Korean army has cut its ground personnel to 520,000, and it is headed toward about
390,000 in 2020. See BruceW. Bennett, “A Brief Analysis of the Republic of Korea’s Defense Reform
Plan,” OP-165-OSD (Santa Monica, Calif.: RAND, 2006), pp. 2–3, 28–35; and Sung-ki Jung, “South
to Boost Surgical Strike Capability against North,” Korea Times, June 26, 2009.
come.15 Second, it could be overthrown and replaced with a new dictatorship
that establishes control over North Korea’s political and military institutions.
Third, as improbable as it seems today, North Korea’s government (perhaps
Kim Jong-il’s successor) could decide that North Korea is a failing state that
should unify peacefully with South Korea on South Korean terms. This socalled
soft landing could be similar to that of East Germany. In each of these
scenarios, the maintenance of control over North Korean society, the military,
and other critical government institutions should prevent signiªcant instability
and thus would not lead to a situation requiring outside intervention.
The fourth possibility is government collapse, in which the Kim family’s authority
is challenged, but no one manages to establish political control, and
North Korea becomes a failed state. The manner in which collapse occurs
could range from relatively benign to highly dangerous. A particularly perilous
collapse scenario would feature multiple powerful political and military
leaders vying for control. North Korea could quickly become a warlord state,
where competing civilian and military leaders claim to rule swaths of the
country and battle one another for control of territory or resources. Collapse
might also occur during wartime, which would be particularly hazardous:
the North Korean military would be deployed, with orders to ªre on Combined
Forces Command troops, and North Korea’s vast military reserves
(almost 8 million people strong) would be mobilized and armed. The war
may have already disrupted food distribution and triggered refugee ºows.
Weapons of mass destruction are likely to have been dispersed and may have
even been used. Therefore a wartime collapse, or a collapse that leads to a
signiªcant power struggle within the country, would create serious instability
in North Korea.
Collapse might also occur in a relatively benign manner. North Korea could
collapse in peacetime—perhaps in a scenario in which Kim Jong-il’s death triggers
a contested succession, and no leader successfully secures power over the
country’s political and military institutions. Although this scenario is still
fraught with uncertainty, at least the North Korean military and security services
would not be on a war footing, and the vast majority of the reserves
would not be mobilized. Additionally, a collapse would be less dangerous if
North Korea’s civilian and military leaders were not jockeying for power and
gathering military units behind them. In this more benign scenario, no one is
clearly in control in Pyongyang; government and military leaders are ºeeing
International Security 36:2 88
15. Byman and Lind, “Pyongyang’s Survival Strategy”; Haggard and Noland, Famine in North Korea;
Noland, “Why North Korea Will Muddle Through”; and McDonald, “Hardships Fail to
Loosen Regime’s Grip in N. Korea.”
the sinking North Korean ship and seeking asylum in China or elsewhere. Military
units and security services are disintegrating, rather than rallying around
particular leaders, and are heading home to their families—similar to the tatters
of Saddam Hussein’s army in 2002. North Korea would be left with a political
vacuum and a leaderless and dissolving military.
Any analysis of a North Korean collapse pivots on the assumptions made
about how that collapse occurs. Such assumptions drive the number and type
of missions that would be necessary to stabilize the country and the expected
difªculty of those missions. For example, pacifying a North Korea run by powerful
warlords would pose additional challenges relative to one in which military
and political power had simply dissipated. And stabilizing a country that
has already been mobilized for war, in which perhaps hundreds of thousands
of refugees have already taken ºight and WMD may have been used, would
require more missions—with a greater number of forces—relative to a peacetime
collapse.
In our analysis, we estimate force requirements for completing the most
essential stabilization missions after a relatively benign collapse scenario.
In other words, we assume that collapse occurs in peacetime, without prior
military mobilization; that North Korean leaders are ºeeing the country or
hiding, rather than preparing for a ªght; and that North Korea’s military does
not offer signiªcant resistance against stability forces.
Many analysts would question the likelihood of this scenario. For example,
they might argue that Pyongyang’s political and military leaders would surround
themselves with arms and men and would resist an intervention force
for fear of harsh punishments after uniªcation. We share the skepticism of
those critics. However, we assume this relatively optimistic scenario not because
it is the most plausible, but because it is the most analytically useful: this
“best-case” collapse scenario demonstrates that even benign assumptions produce
extremely demanding force requirements for stabilizing a collapsed
North Korea. More pessimistic scenario assumptions would increase force requirements
and would lengthen the duration of stabilization missions.
Stabilizing North Korea: Problems, Missions, and Requirements
Baghdad’s descent into chaos in the days after the toppling of Saddam
Hussein’s regime serves as a vivid reminder of the dangers associated with the
collapse of a government and the need for detailed planning before it occurs.16
The Collapse of North Korea 89
16. James Fallows, “Blind into Baghdad,” Atlantic, Vol. 293, No. 1 (January/February 2004),
pp. 52–74; and Michael E. O’Hanlon, “Iraq without a Plan,” Policy Review, No. 128 (December and
A government collapse in North Korea could trigger problems that South
Korea and other countries have a profound interest in mitigating.17 These
problems include the outbreak of starvation and disease in North Korea; mass
refugee ºows across borders; “loose” nukes and other forms of WMD; and the
potential for ongoing insurgency and violence throughout the country (turning
North Korea into another failed state). Many of these problems are related.
For example, mass starvation and disease would be horrifying in and of themselves,
but they could also vastly increase refugee ºows once regime control
mechanisms began to fail. As a consequence, many of the military missions
we discuss address multiple problems, and the missions themselves are often
mutually reinforcing. Ahumanitarian mission, for example, would not only alleviate
starvation; it would also reduce refugee ºows.
Five principal military missions would be necessary to mitigate the problems
discussed above: (1) stability operations, including direct humanitarian
relief and policing of major cities and roads; (2) border control; (3) elimination
of WMD; (4) disarmament of conventional weapons; and (5) deterrence or defeat
of any military resistance.
stability operations
A government collapse could unleash tremendous instability throughout
North Korea. A contestation of power could disrupt the provision of public
services, including the Public Distribution System that supplies North Koreans
with about half of their absolute minimum caloric needs.18 Because of the famine
in the late 1990s, as well as ongoing food shortages, the North Korean peo-
International Security 36:2 90
January 2005). U.S. civilian and military analysts did extensive prewar planning for the Iraqi occupation;
many of their recommendations were rejected by ofªcials in George W. Bush’s administration.
See Nora Bensahel, Olga Oliker, Keith Crane, Richard R. Brennan Jr., Heather S. Gregg,
Thomas Sullivan, and Andrew Rathmell, After Saddam: Prewar Planning and the Occupation of Iraq
(Santa Monica, Calif.: RAND, 2008), pp. 239–240; and Donald P. Wright and Timothy R. Reese, ON
POINT II: Transition to the New Campaign: The United States Army in Operation Iraqi Freedom, May
2003–January 2005 (Ft. Leavenworth, Kans.: Combat Studies Institute, May 26, 2004), pp. 70–80.
17. Previous discussions of these problems include Paul B. Stares and Joel S. Wit, “Preparing for
Sudden Change in North Korea,” Council Special Report, No. 42 (New York: Council on Foreign
Relations, January 2009); Marcus Noland, Avoiding the Apocalypse: The Future of the Two Koreas
(Washington, D.C.: Institute for International Economics, 2000); Jonathan D. Pollack and Chung
Min Lee, Preparing for Korean Uniªcation: Scenarios and Implications (Santa Monica, Calif.: RAND,
1999); Robert D. Kaplan, “When North Korea Falls,” Atlantic (October 2006), http://www
.theatlantic.com/doc/200610/kaplan-korea; Bruce Klingner, “New Leaders, Old Dangers: What
North Korean Succession Means for the U.S.,” Backgrounder, No. 2397 (Washington, D.C.: Heritage
Foundation, April 7, 2010); Robert Collins, “Patterns of Collapse in North Korea,” The Combined
Forces Command C5 Civil Affairs Newsletter, Seoul, Korea, January 1996, pp. 2–12; and
David S. Maxwell, “Catastrophic Collapse of North Korea: Implications for the United States Military”
(Fort Leavenworth, Kans.: United States Army Command and General Staff College, 1996).
18. Haggard and Noland, Famine in North Korea, p. 193.
ple are already severely malnourished. One study estimates that half of North
Korean children are stunted or underweight, while fully two-thirds of young
adults are malnourished or anemic.19 A disruption in the food supply and in
health care services could thus create a humanitarian crisis that rapidly kills
hundreds of thousands of people.
A lack of food combined with uncertainty about who is in power may have
other destabilizing effects. Inadequate food supplies have already encouraged
criminal activity, as evinced by widespread reports of soldiers stealing crops
from farmers.20 The end of central government control could therefore lead to
looting and banditry. As in Iraq, general insecurity could fuel support for insurgency
as people seek the protection of militias—a serious danger in a society
where nearly 40 percent of the population is on active duty or in the
military reserves. Hunger combined with anarchy could lead North Koreans
to seek refuge in other countries.
the mission. Analysts note that no two stability operations are the same:
each demands different kinds of missions, and the difªculty of those missions
varies depending on case-speciªc circumstances.21 In a North Korean stability
operation, military forces would need to secure the lines of communication
that would serve as supply routes for humanitarian aid—major roads, ports,
railroads, and airports. They would also need to escort relief convoys to population
centers for distribution; supplies sent into anarchic areas without protection
could be seized by thieves and diverted to the black market, or looted by
desperate people in search of food, water, and medicine for their families.22 In
addition, stability forces would be tasked with setting up, stafªng, and protecting
food distribution centers.
Astability operation in North Korea would also assume policing activities to
provide public security. North Korea’s police may not be useful, as in the case
of Iraq after the U.S. invasion in 2003: Iraqi police did not show up for work,
and police stations were destroyed by looters.23 Regardless, a stability force
The Collapse of North Korea 91
19. National Intelligence Council (NIC), Strategic Implications of Global Health, ICA 2008-10D
(Washington, D.C.: NIC, December 2008), p. 46.
20. Bradley K. Martin, Under the Loving Care of the Fatherly Leader: North Korea and the Kim Dynasty
(New York: Thomas Dunne, 2004), p. 517.
21. For discussion, see John McGrath, Boots on the Ground: Troop Density in Contingency Operations,
Global War on Terrorism Occasional Paper, No. 16 (Fort Leavenworth, Kans.: Combat Studies Institute
Press, 2006); and U.S. Army, The Infantry Brigade, FM 7–30, change 1 dated October 31, 2000
(Washington, D.C.: U.S. Army, October 3, 1995), table J-7, J-34.
22. Banditry could be a serious problem in North Korea, given defector reports of serious food
shortages within the military—people whose weapons would give them the ability to threaten relief
convoys. See “Former N.K. Soldiers Testify about Worsening Food and Rights Conditions,”
Yonhap News Agency, February 21, 2011.
23. Bensahel et al., After Saddam, pp. 85, 125. Police forces are “the ªrst counterinsurgent organizacould
not rely on North Korean police because they are notoriously corrupt,
their loyalties would be questionable, and they may have no legitimacy in the
eyes of the people, given their repression of the public under the Kim regime.
Therefore stability forces would need to replace the local police, and screen
and train North Korean personnel to gradually take over this function.
mission requirements. Historically, the kinds of forces that have been appropriate
for the tasks described above include light infantry, military police,
engineers for infrastructure construction, and some special forces and others
trained in psychological operations and civil affairs. Special forces would enable
the stability force to communicate better with the local community, increasing
the force security and the efªciency of their mission. For some of these
missions, a stability force could receive support from nonmilitary personnel
(i.e., civilian police forces and military contractors).24
Analysts typically estimate force requirements for stability operations based
on the size of the population to be controlled.25 In his landmark study, James
Quinlivan of RAND calculated that historically the easiest missions, those
with a docile public and little resistance, required 4 soldiers per 1,000 people in
the population. This force ratio is comparable to the density of police under
normal circumstances in major cities in the industrialized world.26 More dangerous
threat environments, however, required a larger number of forces for
every person who must be policed: Quinlivan notes that more difªcult cases
required 20 soldiers per 1,000 inhabitants.27 A subsequent study by John
McGrath argues that, with the sole exception of the NATO peacekeeping operation
in Kosovo in 1999, most occupations and stability operations have fewer
peacekeepers than the 20:1,000 ratio. In the post–World War II occupations of
Japan and Germany, the British occupation in Malaya, and NATO’s mission in
Bosnia, troop deployments were far smaller, closer to the ratios required for
domestic policing. For example, in U.S.-occupied Iraq, the ratio of troops to inhabitants
in Baghdad reached 7 per 1,000 only after the “surge.”28 McGrath
International Security 36:2 92
tion that has to be inªltrated and neutralized.” David Galula, Counterinsurgency Warfare: Theory
and Practice (New York: Praeger, 1964), p. 31.
24. On operations performed by defense contractors in Iraq, for example, see Bensahel et al., After
Saddam, pp. 127, 140.
25. James T. Quinlivan, “Force Requirements in Stability Operations,” Parameters, Vol. 25, No. 4
(Winter 1995–96), pp. 59–69. For discussion of this method, see McGrath, Boots on the Ground, pp. 2,
97; and Peter J.P. Krause, “Troop Levels in Stability Operations: What We Don’t Know,” Audit of
the Conventional Wisdom, 07-02 (Cambridge, Mass.: MIT Center for International Studies, February
2007).
26. Quinlivan, “Force Requirements in Stability Operations”; and McGrath, Boots on the Ground,
chap. 3.
27. Quinlivan, “Force Requirements in Stability Operations,” pp. 62–63.
28. Iraq Index: Tracking Variables of Reconstruction & Security in Post-Saddam Iraq (Washington, D.C.:
concludes that the 20:1,000 ratio is excessive for most cases and recommends a
ratio of 13 per 1,000 for an “average” stability operation. Of this number, he
suggests that at least 30 percent (4 per 1,000) be dedicated to policing.29
A stability operation in North Korea would need to take into account the realities
on the ground (e.g., geography, infrastructure, and population density),
as well as two competing pressures: the need for speed versus the time required
to launch a complicated stability operation. Regarding the need for
speed, the number of starving people would grow daily, causing more chaos,
greater demands on the stability operation, and a more challenging environment
in which the stability forces would have to function. Such worsening
conditions would exacerbate other problems such as refugee ºows, loose
nukes and other WMD, and the likelihood of insurgency. Finally, if instability
were to develop along China’s border, Beijing might decide unilaterally to
send forces into North Korea to stabilize the area—a prospect that ofªcials in
Seoul and Washington regard with great alarm.30 For all of these reasons, a
rapid stabilization of North Korea would be desirable.
A rapid, comprehensive stability operation, however, would probably be
unfeasible for political and logistical reasons. Staffed at McGrath’s ratio of
13 peacekeepers per 1,000 inhabitants, an operation designed to feed and protect
all of North Korea’s 24 million people would require 312,000 personnel.31
Before a force of that magnitude could be moved into North Korea, several political
obstacles would have to be overcome. In particular, Chinese analysts
and ofªcials currently view the insertion of any foreign military forces into
North Korea as an illegal military invasion.32 Policymakers in Beijing, Moscow,
Seoul, and Washington would have to negotiate the numbers, types, and nationalities
of forces to be sent into North Korea; the United States and South
The Collapse of North Korea 93
Brookings Institution, June 28, 2007), p. 7, http://www.brookings.edu/fp/saban/iraq/index.pdf.
McGrath estimates that if one factors in the presence of military contractors, the occupation force
levels were about 10 per 1,000. McGrath, Boots on the Ground, p. 136.
29. For detailed analysis, see McGrath, Boots on the Ground, chap. 4.
30. Stares andWit, “Preparing for Sudden Change in North Korea,” pp. 6–7; Drew Thompson and
Carla Freeman, “Flood across the Border: China’s Disaster Relief Operations and Potential Response
to a North Korean Refugee Crisis” (Washington, D.C.: U.S.-Korea Institute, School of Advanced
International Studies, Johns Hopkins University, April 1, 2009).
31. Population data are from Central Bureau of Statistics, DPR Korea 2008 Population Census, National
Report (Pyongyang: Central Bureau of Statistics, DPR Korea, 2009), pp. 18–22, http://
unstats.un.org/unsd/demographic/sources/census/2010_PHC/North_Korea/Final%20national
%20census%20report.pdf.
32. Interviews by authors, Beijing University, Shanghai Institutes of International Studies, China,
April 2010; and Stares and Wit, “Preparing for Sudden Change in North Korea,” p. 7. For other
views on the legality of such operations in this context, see Shin Beomchul, “A Review of the Legalities
Associated with a Sudden Change in North Korea,”Working Paper, No. 6 (Seoul: Ilmin International
Relations Institute, Korea University, October 2010).
Korea would need to coordinate their response; and diplomats from these or
other countries might lobby the United Nations to issue a mandate for a multilateral
stabilization effort. Overcoming the political obstacles to launching a
military stabilization operation in North Korea could consume a great deal of
time.
The logistical hurdles to a rapid, comprehensive stability operation would
be similarly daunting. The ªrst challenge would be assembling such a massive
force, a task made much more complicated if portions of it would be coming
from outside the region. Moving into the theater massive numbers of personnel,
as well as equipment and humanitarian aid, would be a herculean undertaking.
A force of this size could not be rapidly moved into North Korea; its
meager road network and inadequate airªelds and ports would be overwhelmed.
33 Finally, military planners would most likely reject a strategy of inserting
a large number of troops throughout North Korea without regard to
securing the lines of communication and other infrastructure required for logistical
support. In sum, a rapid, comprehensive stability operation in North
Korea may be both politically and logistically infeasible.
A more likely approach, therefore, would reconcile these political and logistical
realities with the need to provide aid to as many North Koreans as possible,
as quickly as possible. With this in mind, we conceptualize a stability
operation as having two main elements. The ªrst element would be an operation
that would stabilize the country sequentially from south to north (we assume
that, as heir to North Korea, South Korea would play a leading role). To
model this effort, we conceptualize North Korea as divided into ªve tiers (see
ªgure 1). The force would begin by stabilizing only the southernmost tier—
establishing control over the lines of communications, creating relief centers,
replacing police forces, and so forth. We estimate that 13 soldiers per 1,000 of
population (as proposed by McGrath) would be devoted to the active tier. After
the area had been stabilized, the stability force would move northward into
the new tier at a ratio of 13:1,000, leaving behind a smaller troop component
(we estimate 6:1,000).34
Thus, as the stability forces moved northward from tier to tier, overall force
requirements would rise. Based on North Korean demographic data, table 1
displays population ªgures by tier and shows the number of stability forces
that would be required for each phase of the stabilization effort. At its peak,
the operation would require approximately 180,000 soldiers.
International Security 36:2 94
33. Of North Korea’s 25,000-kilometer road network, fewer than 750 kilometers of roads are
paved. Mountains virtually divide the east and west coasts, which are connected by only one major
highway. See CIAWorld Factbook, North Korea, https://www.cia.gov/library/publications/theworld-
factbook/geos/kn.html.
34. We estimate 6:1,000 based on these being largely replacement police forces.
A second element of the stability operation would deliver humanitarian aid
through North Korea’s two major ports ahead of the force advance and set up
coastal lodgments. Amphibious forces would seize, defend, and repair North
Korea’s two major ports (Nampo, in the west, which supplies Pyongyang, and
Chongjin, in the northeast). These forces could then advance into the upper
tiers along the coastal ºanks to deliver humanitarian aid. This approach would
The Collapse of North Korea 95
Figure 1. North Korean Civilian and Military Population by Tier
NOTE: Civilian population data are based on Central Bureau of Statistics, DPR Korea 2008
Population Census (Pyongyang: Central Bureau of Statistics, 2009). Military populations by
tier are authors’ estimates.
suit the realities on the ground, not only because Nampo and Chongjin are the
country’s two major ports,35 but also because the country’s population is concentrated
along the coasts (see ªgure 2).
Calculating the size of this element of the stability operation is difªcult without
knowing speciªc, unavailable details about the ports. Excluding these
details from the analysis, however, does not signiªcantly affect our overall
force estimates: rather it means that our model somewhat underestimates requirements
in the early phase (when additional forces would be needed to
secure the ports and their surrounding areas) and somewhat overstates requirements
in the later phases, because the coastal areas would already have
been stabilized.
This model provides a good estimate of how to conceptualize and calculate
force requirements for a North Korean stability operation, but it also raises
many questions that are currently unanswerable. For example, how long
would it take to stabilize a given tier? Two weeks? A month? Several months?
The answer would inºuence the pace of force requirements. Other issues are
how extensively and how rapidly North Korean personnel could replace outside
stability forces (namely, those forces sized at a ratio of 6:1,000 of population
who remain in the stabilized tiers). Recruiting North Koreans (who are
screened and trained) into the police force would be desirable not only to reduce
the burden on outside countries, but also to increase the likelihood of a
successful operation.36 Yet, it could take time before North Koreans could be
brought in, and at this point, how much time remains unknown.37
International Security 36:2 96
35. The ports at Nampo and Chongjin handle more than half of North Korean seaborne shipping.
For data, see “Korea: Port of Pusan,” AsiaTradeHub.com, http://www.asiatradehub.com/n.korea/
ports.asp.
36. On the issue of peacekeeper ethnicity and mission success, see Jason Lyall, “Are Coethnics
More Effective Counterinsurgents? Evidence from the Second Chechen War,” American Political
Science Review, Vol. 104, No. 1 (February 2010), pp. 1–20.
37. Previous analyses show that it can take a year to develop and ªeld police forces. James Dob-
Table 1. Force Requirements for a Sequenced Stability Operation in North Korea (per tier,
per phase)
Phase
Initial Second Third Fourth Fifth Final
5 0 0 0 0 69,300 32,000
4 0 0 0 79,200 36,500 36,500
Tier 3 0 0 107,500 49,600 49,600 49,600
2 0 32,400 14,900 14,900 14,900 14,900
1 24,800 11,500 11,500 11,500 11,500 11,500
Totals 24,800 43,900 133,900 155,200 181,800 144,500
border control
In the event of government collapse in North Korea, the cessation of
government-provided food and services, as well as rising internal violence,
may encourage many North Koreans to ºee to neighboring countries.38 These
countries would consider using military forces to secure their borders.
North Korea’s population is concentrated around Pyongyang and along
each coast. Some refugees could head for South Korea (approximately 150 kilometers
from Pyongyang). The terrain in that direction is not as difªcult as the
rough mountain terrain that lies northward, but the mountains on the eastern
half of the peninsula still pose signiªcant barriers. Other refugees, particularly
among the more than 11 million people who live above the peninsula’s “narrow
neck” could head north. North Korea shares a 19-kilometer border with
Russia at its far northeastern corner, along the Tumen River; otherwise, most of
its northern border is shared with China. The Chinese province of Jilin contains
the Yanbian Korean Autonomous Prefecture, which has a population of
more than 500,000 ethnic Koreans (40 percent of its population). During the
1995–97 famine in North Korea, an estimated 400,000 North Koreans crossed
into China in search of food. The United States and the UN estimate that
30,000–50,000 North Korean refugees live in China.39 Importantly, these North
Koreans sought refuge despite the risk of imprisonment, torture, or execution
if caught and sent home.40 In the absence of such deterrents, the ºow of refugees
would likely be much higher.41
A refugee crisis on the Korean Peninsula could create signiªcant regional instability
and possibly require countries to provide food, security, housing, and
other care to ºeeing refugees. The Chinese fear that North Korean soldiers
might cross into China, bringing their weapons with them and engaging in
banditry and other violent activity.42 As noted, refugee activity on the border
The Collapse of North Korea 97
bins, Andrew Rathmell, Keith Crane, Seth G. Jones, and John G. McGinn, America’s Role in Nation
Building: From Germany to Iraq (Santa Monica, Calif.: RAND, 2003), pp. 151–152. On Germany’s
training and integration of former East German Nationale Volksarmee soldiers, see Dale
Herspring, Requiem for an Army: The Demise of the East German Military (Lanham, Md.: Rowman
and Littleªeld, 1998), chap. 6.
38. Na Jeong-ju, “3 Million NK Refugees Expected in Crisis: BOK,” Korea Times, January 26, 2007.
39. For estimates during the famine, see Andrew Natsios, “The Politics of Famine in North Korea”
(Washington, D.C.: United States Institute of Peace Press, August 2, 1999), p. 11. For a 2006 estimate,
see Rhoda Margesson, Emma Chanlett-Avery, and Andorra Bruno, “North Korean Refugees
in China and Human Rights Issues,” CRS Report for Congress (Washington, D.C.: Congressional
Research Service, Library of Congress, September 26, 2007), Order Code RL34189, p. 4.
40. On the tragedy of North Korean refugees in China, see Human RightsWatch, The Invisible Exodus:
North Koreans in the People’s Republic of China (New York: Human Rights Watch, November 19,
2002).
41. Some North Koreans have relatives in Japan. Although a seaborne refugee problem may necessitate
some maritime stability operations, this analysis does not incorporate this potential aspect
of the refugee problem.
42. Interviews by authors, Beijing University, Shanghai Institutes of International Studies, China.
could spur China to send the PLA into North Korea to stabilize the area. If the
United States and South Korea also sent forces into North Korea to help with
the refugee problem, then the intermingling of Chinese, Korean, and American
soldiers could lead to confusion, skirmishes, and escalation.
the mission. In addition to the stability operation forces described above,
North Korea’s neighbors may seek to contain a refugee ºow through border
control operations. Guards would be stationed at various points along North
International Security 36:2 98
Figure 2. North Korea’s Population Density
SOURCE: Perry-Castañeda Library map collection, http://www.lib.utexas.edu/maps/middle
_east_and_asia/north_korea_pop_1972.jpg.
Korea’s borders, tasked with observing people who approach and intercepting
and apprehending those attempting to cross. Border guards would coordinate
with peacekeepers from the stability operation to direct and transport refugees
to refugee camps or aid distribution centers. (We assume that the personnel
required to operate such camps would be part of the stability operation
described above.)
mission requirements. To estimate the number of guards needed to contain
a North Korean refugee ºow, we draw on historical cases of border
control. One example involves the 3,169-kilometer U.S.-Mexico border, which
continues to experience a huge amount of illegal trafªc. This lightly patrolled
border has approximately 3 agents per kilometer on the U.S. side.43 Many
Mexican migrants and smugglers of contraband trade, including drugs, cross
this notoriously porous border illegally.44
On different occasions, efforts to increase the number of agents on the U.S.
side of the border have deterred illegal crossings. In Operation Blockade,
for example, in 1993, 400 border guards manned a 32-kilometer segment of
the U.S.-Mexican border near El Paso, Texas, raising the number of border
guards to 12 per kilometer. This effort was heralded as a dramatic success:
a Department of Justice report concluded that the operation “stopped numerous
day-crossers, resulting in a 70 percent drop in El Paso Sector apprehensions.”
The report noted that migrants and smugglers sought instead to cross
at more lightly patrolled sections of the border: “They no longer came through
central El Paso.”45 The program’s success led other cities to adopt its methods
and prompted a change in strategy by the U.S. Border Patrol.46
A RAND study of the Vietnam War, which examined the problem of illegal
crossings and smuggling between North and South Vietnam, recommended a
higher number of border guards. This study proposed 22 agents per kilometer
for containing a population under peacetime conditions.47 In non-peacetime
The Collapse of North Korea 99
43. In 2003 there were 9,633 agents along the 3169-kilometer border. See “The U.S. Border Patrol:
Failure of the Administration to Deliver a Comprehensive Land Border Strategy Leaves Our Nation’s
Borders Vulnerable,” prepared for Rep. Bonnie G. Thompson, Minority Staff of the House
Committee on Homeland Security, U.S. House of Representatives, May 2005, p. 3.
44. In 2003 the United States intercepted 900,000 migrants from Mexico. U.S. border ofªcials estimate
that for every one of these arrests, four people successfully crossed the border illegally. See K.
Jack Riley, “Border Control,” in Infrastructure, Safety and Environment (Santa Monica, Calif.: RAND,
2006), p. 602.
45. United States Department of Justice, “Operation Gatekeeper: An Investigation into Allegations
of Fraud and Misconduct,” July 1998, http://www.usdoj.gov/oig/special/9807/gkp01.htm.
46. Douglas S. Massey, “Borderline Madness: America’s Counterproductive Immigration Policy,”
in Carol M. Swain, Debating Immigration (Cambridge: Cambridge University Press, 2007), p. 131.
47. C.V. Sturdevant, The Border Control Problem in South Vietnam (Santa Monica, Calif.: RAND, June
1964), p. 55.
situations, and in situations where essentially zero porosity is tolerated, the
number of agents would need to be much higher.48
From these cases, we adopt a midrange estimate of 17 guards per border kilometer
and calculate the requirements for stabilizing the Chinese and Russian
borders. (We assume that, initially, the current South Korean military presence
along the demilitarized zone [DMZ], not to mention land mines and other obstacles,
would be sufªcient to prevent refugees from entering South Korea).49
This metric generates a requirement of 323 guards along the 19-kilometer
Russian border and 24,072 guards along the 1,416-kilometer Chinese border,
totaling approximately 24,400 soldiers for this mission.
Our analysis assumes that a stability operation is being performed, making
large-scale refugee ºows much less likely. If, however, an extremely large refugee
ºow develops, then a more substantial force (and reinforcements along the
DMZ) would likely be required.
elimination of wmd
In the wake of a government collapse in North Korea, many policymakers in
Beijing, Seoul, Washington, and elsewhere would consider securing North
Korea’s WMD arsenal a vital mission. As noted in the 2010 U.S. Quadrennial
Defense Review, “The instability or collapse of a WMD-armed state is among
our most troubling concerns. Such an occurrence could lead to rapid proliferation
of WMD material, weapons, and technology, and could quickly become a
global crisis posing a direct physical threat to the United States and all other
nations.”50 The dispersal of the North Korean arsenal could result in assembled
atomic bombs, loose ªssile material, pathogens, and toxic chemicals
reaching the global black market. This risk is elevated because to subsist in
North Korea today, many ofªcials and others already engage in extensive
black market activity.51
North Korea has a substantial WMD program. It is thought to have produced
enough ªssile material for perhaps ªve to twelve nuclear weapons.52
International Security 36:2 100
48. Sturdevant provides data on tense borders, for example, in Berlin (staffed at 176 guards per kilometer).
Ibid., p. 55.
49. If these forces were sent into North Korea for stabilization missions, additional forces would
need to be assigned to control the South Korean border. The North-South border is 248 kilometers,
which (using the 17 guards per kilometer ratio) generates a requirement of 4,216 guards.
50. U.S. Department of Defense, Quadrennial Defense Review Report (Washington, D.C.: Department
of Defense, February 2010), p. iv; and U.S. Department of the Army, “Chemical, Biological, Radiological,
Nuclear, and High Yield Explosives Operational Headquarters,” Field Manual Interim,
No. 3-90.10 (Washington, D.C.: U.S. Department of the Army, January 2008).
51. Sheena Chestnut, “Illicit Activity and Proliferation: North Korean Smuggling Networks,” International
Security, Vol. 32, No. 1 (Summer 2007), pp. 80–111.
52. David Albright and Paul Brannan, “The North Korean Plutonium Stock” (Washington, D.C:
While the data on North Korea’s nuclear weapons are uncertain, even less is
known about its chemical and biological weapons programs. According to one
report, “Estimates of North Korea’s total stockpile vary by more than an order
of magnitude,” with estimates of chemical stockpiles ranging from about 180
to 250 tons to 2,500 to 5,000 tons.53 North Korea’s WMD program is spread
across numerous facilities, both known and covert. In 2009 South Korea’s defense
minister testiªed that there were about 100 sites in North Korea related
to its nuclear program.54 We assume roughly the same number of facilities for
its chemical and biological weapons programs (i.e., a total number of WMD facilities
of perhaps 200).
North Korea also has a cadre of scientists and engineers who developed its
WMD programs, and it likely has many sensitive documents that recorded
what these scientists learned in their research and testing. Following a government
collapse, North Korean scientists and engineers would likely be worried
about food, money, and safety (for themselves and their families), so they may
be lured by opportunities abroad to sell their WMD knowledge to terrorist organizations
or countries seeking to develop nuclear and other weapons. Thus,
following a collapse, a vital mission would be to secure North Korea’s WMD
program in all of its various forms.
the mission. In the event of a North Korean government collapse, the elimination
of WMD would be not only one of the most important missions, but
probably the most challenging. Most of North Korea’s critical WMD facilities
are located north of Pyongyang: reaching these facilities would take a great
deal of time, because stability forces would ªrst need to secure the lines of
communication and move northward through the mountainous northern region
with its extremely poor road networks. As difªcult to accept as it may be,
foreign leaders may have to wait weeks or even months before stabilization
forces could secure and inspect most North Korean WMD facilities. The good
news, however, is that Korea is a peninsula: countries could attempt to contain
weapons, ªssile material, and WMD personnel by sealing off North Korea’s
coastline and its borders (another important reason to perform a border control
operation).
A mission to eliminate North Korea’s WMD would have four key components,
which would need to be conducted simultaneously. The ªrst task would
The Collapse of North Korea 101
Institute for Science and International Security, February 20, 2007), http://www.isis-online.org/
publications/dprk/DPRKplutoniumFEB.pdf.
53. “North Korea: Chemical Weapons Program,” GlobalSecurity.org, http://www.globalsecurity
.org/wmd/world/dprk/cw.htm (accessed on October 23, 2009).
54. “Seoul Suspects about 100 Sites in N.K. Linked to Nuclear Program,” Korea Herald, October 5,
2009.
be to contain the problem, which would require control of North Korean ports,
naval/coast guard interception of vessels leaving North Korean waters, aerial
reconnaissance of air trafªc and air interception of aircraft leaving North
Korea, and cooperative efforts with North Korea’s neighbors—China, Russia,
and South Korea—to prevent overland escape of weapons, ªssile material, and
personnel. These operations would need to continue until all North Korean
WMD facilities and potential dispersal sites had been searched and secured.
A second component of the mission would be to conduct surveillance and
intercept targets of opportunity. Reconnaissance assets and special forces
would be sent ahead of “friendly” forces to observe suspected WMD facilities
in northern North Korea. If these forces detected suspicious activities suggesting
that WMD matériel or weapons were being looted or stolen, assault teams
could be dispatched to stop those activities and contain the materials being
moved. These teams could be inserted through the secured ports or from ships
on the North Korean coast, including army, Marine, and special forces.
A third component of the mission would be to secure the highest priority
WMD targets. This mission would be conducted in tandem with the principal
stability operation moving northward tier-by-tier through North Korea. Units
would seize, secure, and search suspected WMD sites within the “active” tier
and then proceed into the next tier.
The fourth component of the mission would involve a sequenced, methodical
sweep through North Korea to secure and inspect all possible WMD
facilities, including potentially thousands of underground facilities.55 This
effort would consolidate WMD at central storage sites, where specialized personnel
would be called in to evaluate them, a process usually referred to as
“exploitation.”56 International scientists and weapons inspectors, including
personnel from organizations such as the International Atomic Energy Agency,
would be part of this group. They would want to study North Korea’s technology
and determine which countries or groups provided technical assistance.
They would also want to gather vital scientiªc documentation and construct a
personnel chart—a “deck of cards” of the key players in North Korea’s WMD
program,57 who would need to be located and prevented from leaving the
country. Once the examination of these materials was complete, Koreans
International Security 36:2 102
55. Barbara Demick, “Vision on Tunnels Drives N. Korean Defense,” Boston Globe, November 28,
2003.
56. See U.S. Department of the Army, “Chemical, Biological, Radiological, Nuclear, and High
Yield Explosives Operational Headquarters,” p. 4-4.
57. During Operation Iraqi Freedom in 2003, the United States identiªed a “card deck” of senior
Iraqi personnel that it sought to capture and interrogate. A similar approach might be taken in
North Korea.
would need to make decisions about the elimination of the conªscated materials.
During this entire process, new intelligence about previously unknown
sites would be gathered from captured records and personnel, and shared with
WMD elimination teams and forces involved in conventional disarmament
efforts.
mission requirements. Naval and air forces would mainly resource the
ªrst component of this mission, in addition to the border control personnel described
earlier. The second and fourth components would be performed as
part of the stability operation or conventional disarmament operation. Thus
we assume that the third component (seizure of high-priority facilities) is the
key component for estimating necessary forces for the WMD elimination
mission.
Securing each of North Korea’s high-priority WMD facilities would be akin
to launching a “raid”: a forced-entry operation into a facility, which usually involves
surprising and overcoming local defenders. Raids against installations
with light to modest defenses typically involve 100–200 Special Forces soldiers;
increasing the size beyond these numbers vastly increases the complexity. The
raid of Osama bin Laden’s compound in Abbottabad, Pakistan, was performed
by 79 U.S. commandoes in four helicopters.58 The Israeli raid in 1976 to
free hostages held at Entebbe Airport involved between 100 and 200 commandoes
in the ground task force. The United States sent 56 special operations soldiers
to storm the Son Tay prisoner of war camp in North Vietnam in 1970.59
For the failed U.S. operation in 1980 to liberate U.S. hostages held in Iran, the
United States deployed an assault force of 132 men. In Somalia, the U.S. raid to
capture warlord Mohamed Farrah Aideed and other clan leaders involved 160
soldiers, including Army Rangers and Delta Force operators.60 With the exception
of the Entebbe mission, all of the raids described above relied primarily on
helicopter-borne forces.
Using these historical cases, we estimate that securing and searching a major
WMD facility would require roughly 200 soldiers, or about two companies.
This estimate is based on our assumption that the operation would face a low
level of resistance from North Korean forces. Without North Korean military
The Collapse of North Korea 103
58. Steven Lee Myers and Elisabeth Bumiller, “Obama Calls World ‘Safer’ after Pakistan Raid,”
New York Times, May 2, 2011.
59. On Entebbe, see Raymond Carroll, “How the Israelis Pulled It Off,” Newsweek, July 19, 1976,
p. 42; and Robert M. Smith, “Raid on a P.O.W. Camp: Build-up—and Letdown,” New York Times,
November 27, 1970. The prisoners of war had been moved from the camp.
60. On the hostage rescue mission, see Mark Bowden, “The Desert One Debacle,” Atlantic, Vol.
297, No. 4 (May 2006), pp. 62–77, http://iran.theatlantic.com/interactive_article_page_1.html. On
Mogadishu, see Mark Bowden, Black Hawk Down: A Story of Modern War (Berkeley, Calif.: Atlantic
Monthly Press, 1999), p. 5.
passivity, a battalion or a brigade (700–3,500 personnel) would likely be required.
North Korea’s WMD sites are among the country’s most sensitive
and valuable assets; many of them are large facilities and would be heavily
guarded.
As noted above, we assume that North Korea has about 200 WMD facilities.
WMD elimination teams could enter North Korea along with stability forces
(either from the south or from ports). This mission would require approximately
10,000 troops for tiers 3, 4, and 5 (200 troops for each of 50 sites). If sites
can be prioritized and sequenced, a force of perhaps 3,000–4,000 might sufªce.
Therefore we estimate 3,000–10,000 personnel would be needed for the WMD
elimination mission. If stabilization forces uncover as many tunnels and underground
facilities as have been rumored to exist in North Korea, and if
WMD are found in many of these facilities, this operation could vastly increase
in complexity and duration.
Analysts should understand two realities about a WMD elimination mission
in a collapsed North Korea. First, this effort would be extremely difªcult, particularly
because North Korea has revealed the existence of a uranium-based
nuclear program.61 The location of many of the facilities associated with this
program may not be known, and some of its technologies (such as for centrifuge
production) could be easier than other WMD technology to smuggle out
of the country. Second, China, South Korea, and the United States would view
the recovery of North Korean WMD as a vital mission, and thus would be
likely to send in military forces to accomplish this objective. A shared goal of
securing WMD thus holds signiªcant potential for misperception, conºict, and
escalation among these countries in the aftermath of a North Korean collapse.
disarmament of conventional weapons
In the wake of a government collapse, or at the end of a civil war, looters frequently
raid arms caches, and soldiers frequently disappear with their weapons.
Both activities raise the risk not only of insurgency, but also of banditry
and crimes such as murder and armed robbery.62 They endanger civilians and
particularly children, who could be wounded or killed as they come across
abandoned or hidden armaments or ammunition.63 In numerous cases of civil
International Security 36:2 104
61. Siegfried S. Hecker, “A Return Trip to North Korea’s Yongbyon Nuclear Complex” (Stanford,
Calif.: Center for International Security and Cooperation, Stanford University, November 20,
2010), http://iis-db.stanford.edu/pubs/23035/HeckerYongbyon.pdf. See also “Seoul, Washington
Suspect More N. Korean Uranium Sites,” Chosun Ilbo, December 14, 2010.
62. Daniel Byman, “Understanding Proto-Insurgencies,” Journal of Strategic Studies, Vol. 31, No. 2
(April 2008), pp. 165–200; and Paul Collier, “Rebellion as a Quasi-Criminal Activity,” Journal of
Conºict Resolution, Vol. 44, No. 6 (December 2000), pp. 839–853.
63. David Rohde, “Danger Lingers in Pieces of an Abandoned Arsenal,” New York Times, May 5,
2003.
war, inadequate or nonexistent disarmament efforts contributed to the unraveling
of peace accords and the resumption of hostilities.64 After the toppling of
Saddam Hussein, Iraq’s extensive arms caches were looted, Iraqi soldiers
melted away from their garrisons with their weapons, and disarmament efforts
were inadequate. “There were just not enough boots on the ground, and
the military didn’t give it a high enough priority to stop the looting,” lamented
David Kay, the former chief UN weapons inspector. “Tens of thousands of tons
of ammunition were being looted, and that is what [fueled] the insurgency.”65
Disarmament operations can be performed at various stages after a government
collapse or civil war, but rapid disarmament is inªnitely preferable. In
“Phase I” disarmament operations, or disarmament by command, combatants
turn in their weapons in exchange for amnesty, ofªcial documentation, or
other assistance that will help them transition into either the new security services
or civilian life, or some combination of these. By contrast, in “Phase II”
disarmament operations, a peacekeeping force collects arms voluntarily
turned in by the general public. As Sami Faltas, Glenn MacDonald, and
Camilla Waszink write, however, “Experience clearly points to the need to arrange
an orderly ‘farewell to arms’ very soon after a peace settlement. Their
later removal from circulation is much more difªcult.”66 Such operations are
expensive, because people must be offered a ªnancial incentive to surrender
their weapons, and of dubious effectiveness, because one cannot be certain
how many arms remain uncollected. Most important, as the Iraq experience
shows, a failure to perform disarmament operations early on may trigger years
of civil war or insurgency, resulting in the deaths of many people, increased
cost and duration of the stability operation, and delay in the country’s political
development.
In the wake of a government collapse, the failure to disarm North Korea’s
vast security apparatus would raise the risk of insurgency, banditry, and other
criminal activity. Already, many North Korean police and members of the military
engage in commercial activities and smuggling, some participating in the
The Collapse of North Korea 105
64. For discussion of historical cases, see João Gomes Porto, Chris Alden, and Imogen Parsons,
From Soldiers to Citizens: Demilitarization of Conºict and Society (Aldershot, U.K.: Ashgate, 2007); and
Sami Faltas, Glenn McDonald, and Camilla Waszink, “Removing Small Arms from Society: A Review
of Weapons Collection and Destruction Programmes,” Occasional Paper, No. 2 (Geneva,
Switzerland: Small Arms Survey, July 2001).
65. Kit R. Roane and Edward T. Pound, “A Mess of Missing Ordnance,” U.S. News and World Report,
November 8, 2004, p. 32. A Defense Intelligence Agency report in November 2003 conªrmed
that the vast majority of explosives and ordnance used by insurgents in improvised explosive devices
came from looted Iraqi ammunition caches. Ibid. See also James Glanz and William J. Broad,
“Looting at Weapons Plants Was Systematic, Iraqi Says,” New York Times, March 13, 2005; and
Thomas E. Ricks, Fiasco: The American Military Adventure in Iraq (New York: Penguin, 2007),
pp. 159–162, 191.
66. Faltas, McDonald, and Waszink, “Removing Small Arms from Society,” p. 7.
black market and organized crime.67 Therefore, following regime collapse,
North Korea’s huge security services—military, internal security, police, and
reserves—should be promptly disarmed.
Military planners often think about disarmament as part of a broader set of
operations known as “DDR,” for disarmament, demobilization, and reintegration.
DDR operations disarm a country’s military, release its soldiers from the
military, and provide them with jobs to help reintegrate them into civil society.
It is debatable whether rapid demobilization of the military promotes stability
after regime collapse: the George W. Bush administration’s disbanding of the
Iraqi military, for example, is generally viewed as having been disastrous: fueling
the bloody insurgency that followed the fall of Saddam Hussein.68
Whether or not the North Korean active-duty military should be immediately
demobilized after regime collapse, and how its personnel reintegrated into society
(e.g., put to work in government-sponsored civil construction projects)
are questions of political-military strategy and beyond the scope of this article.
If planners chose to rapidly demobilize the North Korean military, then the
forces necessary to perform that function should be added to the estimates
presented above.
the mission. In a mission to disarm North Korea’s massive security apparatus,
the various forces (military, paramilitary, and police) would ªrst be ordered
to stay in garrison.69 There, stability forces working with North Korean
ofªcers would disarm individual soldiers and round up heavy weapons at
each base (tanks, infantry ªghting vehicles, artillery, and so forth). The stability
forces would need to guard weapons and ordnance located at the bases and
consolidate them at central storage depots.
In addition, stability forces would need to locate and secure a potentially
vast number of additional weapons and ammunition caches. North Korea is
said to have thousands of tunnels and underground facilities, which would
need to be searched for conventional weapons, ammunition stocks, and WMD.
Satellite imagery and other intelligence would direct disarmament teams to
known sites; the North Korean security services would help to identity addi-
International Security 36:2 106
67. John S. Park, “North Korea, Inc.: Gaining Insights into North Korean Regime Stability from
Recent Commercial Activities,” Working Paper (Washington, D.C.: United States Institute of
Peace, April 22, 2009); and Chestnut, “Illicit Activity and Proliferation.”
68. Ricks, Fiasco, pp. 159–162; Larry Diamond, Squandered Victory: The American Occupation and the
Bungled Effort to Bring Democracy to Iraq (New York: Macmillan, 2005), p. 294; Fallows, “Blind into
Baghdad”; and L. Paul Bremer III, “How I Didn’t Dismantle Iraq’s Army,” New York Times, September
6, 2007.
69. For Iraq, the National Security Council and military planners in “Task Force IV” recommended
both keeping soldiers in garrison and eventually using the units in infrastructure reconstruction
operations. See Bensahel et al., After Saddam, pp. 36, 43.
tional facilities. The sites would then need to be secured and inventoried,
and their weapons and munitions transported to central storage locations
where they would be guarded and later assessed.
At the weapons assessment stage (which could take place later), specialized
military and scientiªc personnel from the stability force, along with advisers
from the North Korean military, would determine what to do with the weapons.
Weapons tested and deemed nonfunctional would be destroyed. Otherwise,
as shown by the experience of German uniªcation, the Koreans could
choose to retain weapons and equipment for use by the future military; donate
equipment to international relief agencies or developing nations; transfer
weapons to the United States and any other friendly countries for research
purposes; or render them harmless for use in national commemorations.70
Weapons can also be exported for sale; as in the East German case, the proceeds
could help with the costs of the disarmament operation. In the end, most
East German weaponry was deemed unusable. In far poorer North Korea,
years of deprivation have no doubt led to poor maintenance of what was already
antiquated weaponry. Therefore the bulk of North Korean matériel
would probably be donated or destroyed.
mission requirements. To estimate force requirements for disarmament
operations, we adopt a metric from the United Nations Mission in Cambodia
(UNTAC). In Cambodia, following years of war and genocide, a UN peacekeeping
operation included among its missions the disarming of 200,000 soldiers
and 250,000 militia members. Cambodia had large numbers of diverse
forces (internal security and military forces as well as militias), vast amounts
of weaponry, and ubiquitous land mines. In comparison to many DDR operations,
which often disarm much smaller numbers of soldiers armed with light
weapons, the scale of the Cambodian operation (though still much smaller
than would be a Korean operation) approximates better than most the challenges
of disarming North Korea.
UNTAC’s military component of 16,000 soldiers was tasked with verifying
the withdrawal of foreign forces; supervising the regroupment, cantonment,
and disarmament of 450,000 military and security forces; and collecting and
controlling weapons.71 Additionally, UNTAC forces cleared mines, trained
The Collapse of North Korea 107
70. On the German experience, see Rick Atkinson, “E. German Army Meets Its End,” Washington
Post, February 5, 1994; and Otfried Nassauer, “Surplus: The NVA’s Heritage,” in Edward J. Laurance
and Herbert Wulf, eds., Coping with Surplus Weapons: A Priority for Defense Conversion Research,
BICC Brief No. 3 (Bonn: Bonn International Center for Defense Conversion, June 1995),
p. 37.
71. UNTAC also performed general security and other humanitarian missions that we have analyzed
separately in this article. To avoid double counting, we exclude UNTAC forces that were
5,000 local personnel for mine removal, and conducted mine awareness
and safety programs among the population.72 In UNTAC, the ratio of personnel
devoted to disarmament relative to the number of combatants was assigned
at 35:1,000.
North Korea’s various types of military and security personnel number
more than 1.4 million active-duty personnel (see table 2). Staffed at the
UNTAC ratio (35:1,000), the mission to disarm North Korea’s security personnel
would require 49,000 soldiers.73
Comparison to the German case shows this force estimate to be plausible.
As West Germany began evaluating East Germany’s Nationale Volksarmee
(NVA) arsenal, it encountered staggering manpower demands simply to guard
the massive numbers of arms and ammunition depots. Understaffed planners
scrambled to ªnd the thousands of men to guard the NVA arsenal. The
Bundeswehr’s regular training stalled, as many soldiers—including ofªcers
up to the rank of captain—found most of their time consumed by guard duty.74
If the UNTAC force ratio were used to size the force for a conventional disarmament
mission in North Korea, then—judging by the experience of the
International Security 36:2 108
tasked with other duties. See United Nations, “Cambodia-UNTAC: Background,” http://www.un
.org/Depts/dpko/dpko/co_mission/untacbackgr2.html. See also Grant Curtis, “Transition to
What? Cambodia, UNTAC, and the Peace Process,” Discussion Paper, No. 48 (Geneva, Switzerland:
UN Research Institute for Social Development, November 1993).
72. United Nations, “Cambodia-UNTAC: Background.” Regarding North Korea’s mineªelds, the
North Koreans are reported to have laid more than a million “dumb” land mines along its DMZ
border. (“Dumb” land mines are those that cannot be disarmed and are encased in plastic, which
makes them hard to detect.) Rachel Stohl, “Landmines Remain Issue in Korea,” Defense Monitor,
Vol. 29, No. 5 (June 2000), pp. 1–3.
73. This number of military and security personnel does not include the 7.7 million North Korean
reserve forces; under our assumptions, they remain largely unmobilized. We assume that reserve
weapons caches will be discovered, guarded, and evaluated by the demilitarization force described
here.
74. Herspring, Requiem for an Army, p. 153.
Table 2. North Korea’s Armed Forces
Type Number of Members
Army 1,020,000
Navy 60,000
Air Force 110,000
Paramilitary 190,000
Police 50,000
Total 1,430,000
SOURCES: Ministry of National Defense, Defense White Paper, 2008 (Seoul: Republic of Korea,
March 23, 2008), p. 316; and The Military Balance 2009 (London: International Institute of
Strategic Studies), p. 396.
Cambodian case—this mission could be completed in about two years. If,
however, a massive network of tunnels and underground facilities were discovered
in North Korea, these would need to be searched for conventional
weapons and ammunition (and if such stocks were discovered, they would
need to be guarded). The scale of the East German case—a massive conventional
army with vast numbers of weapons and ammunition caches—thus perhaps
better approximates the demands of disarming North Korea, and
disarmament of the NVA took ªve years.
deterrence or defeat of military resistance
Government collapse in North Korea could unleash insurgency and banditry.
We assume that the North Korean People’s Army would dissolve or cooperate
with a stabilization force; moreover, we argue that after a period of screening
and training, some North Korean soldiers could assist in many of the stability
missions described here—policing, guarding of weapons caches, and so forth.
The Iraq experience shows, however, the risks of military planning that assumes
no resistance. In North Korea, individual soldiers or even some units
could refuse to report to their garrisons for disarmament. Soldiers could refuse
to turn in their weapons because they felt they needed them to feed and protect
their families and neighbors. Soldiers or units could also keep their weapons
to engage in predatory behavior: to intercept humanitarian aid ºowing
into the North and sell it on the black market. In sum, a post-collapse North
Korea would be a highly uncertain environment, and planning for this environment
should not overlook the potential for resistance.
the mission. Military planners would need to allocate combat troops to
confront the potential challenges discussed above. Arapid reaction force based
throughout North Korea—a mix of armor, mechanized infantry, and light aviation
capabilities—could quickly move into areas experiencing resistance or
other kinds of interference with stability efforts.
mission requirements. Two or three brigades added to the stabilization
forces should be adequate for combating limited resistance, particularly if one
assumes that most of these forces would be from South Korea’s well-trained
military. In Operation Desert Storm, a four-brigade U.S. force (the Third Infantry
Division and a single armored cavalry regiment) spearheaded the attack
against an organized defense, which is much more resistance than we assume
in our scenario. Because North Korea’s forces are far less well equipped than
were members of the Iraqi Republican Guard, the highly skilled South Korean
forces would likely be able to meet much of this requirement if they were
properly organized and trained.75 South Korean maneuver brigades usually
The Collapse of North Korea 109
75. More than a decade ago, before famine and years of energy shortages, analysts highlighted
consist of 2,500 to 3,000 personnel, though to be capable of conducting independent
operations, they would need to be augmented with logistical support
and other assets, increasing their size to perhaps 3,500, which would increase
the overall requirement to 7,000–10,500 troops.
critique and discussion
The missions outlined above would generate a substantial troop requirement
for stability operations in a post-collapse North Korea. We estimate that
260,000–400,000 soldiers could be necessary for stability operations and limited
combat operations in this benign scenario (see table 3). Military planners
would most likely want to have additional forces readily available for dealing
with more challenging scenarios, unexpected circumstances, or protracted
commitments.
Some analysts might criticize our analysis as being based on overly optimistic
assumptions about a North Korean collapse. They might argue that it is
highly likely that North Korean political or military elites would seek weapons
and soldiers for protection and ªercely resist stability forces. North Korean
elites may fear prosecution, imprisonment, or execution by South Korea for violations
of human rights or acts committed against South Koreans during or
since the Korean War.76 North Korean ofªcials might also be expected to fear
International Security 36:2 110
North Korea’s conventional inferiority. See Michael O’Hanlon, “Stopping a North Korean Invasion:
Why Defending South Korea Is Easier than the Pentagon Thinks,” International Security, Vol.
22, No. 4 (Spring 1998), pp. 135–170; and Nick Beldecos and Eric Heginbotham, “The Conventional
Military Balance in Korea,” Breakthroughs, Spring 1995, pp. 1–9.
76. Such events include Pyongyang’s 1983 attempted assassination of President Chun Doo Hwan,
which killed 16 and wounded 15 other high-level South Korean ofªcials; the 1987 terrorist bombing
of a KAL airliner, which killed 115 people; the sinking of the South Korean warship Cheonan in
March 2010, which killed 46 sailors; and the shelling of Yongpyeong Island in 2010.
Table 3. Summary of Force Requirements by Mission in North Korea
Mission Requirements (number of soldiers)
Stability operation:
Humanitarian relief and policing
180,000–312,000
(sequenced vs. simultaneous)
Border control 24,000
WMD elimination 3,000–10,000
Conventional disarmament 49,000
Deterrence/Defeat of resistance 7,000–10,500
Total 263,000–405,500
uniªcation because of criminal convictions, imprisonment, and even death
sentences imposed by South Korea on its own former presidents found guilty
of human rights abuses while in ofªce.77 Another cautionary tale for North
Koreans involves the fate of “rogue” leaders such as Saddam Hussein, who
were sentenced to death after falling into the custody of the international community.
North Korea’s high-level military leaders may reasonably perceive
that they have a life-or-death motive to resist uniªcation.
One can easily speculate about the many reasons why a North Korean collapse
could be far more dangerous than this article assumes. Regardless, this
analysis provides an important contribution to policy debates on this subject.
It tells optimists that even if a collapse occurs in a relatively peaceful manner,
stabilizing North Korea would likely require a vast relief effort, involving not
only international aid agencies but also hundreds of thousands of military personnel.
The analysis also has important implications for pessimists: if analysts
believe that a North Korean collapse would be far messier and more dangerous
than we assume, then they should expect that stabilizing North Korea
would require tens of thousands more soldiers than we estimate here. For example,
if one assumes a more dangerous collapse scenario, one could increase
the ratio of stability force personnel per thousand of population from 13:1,000
(the ratio we used) to perhaps 15:1,000 or more, which would signiªcantly expand
the requirements of a stability operation. And assuming that a full-blown
insurgency did develop, subduing it would require far more than the roughly
10,000 troops that we assumed would be needed to quash sporadic resistance.
Some analysts might alternately argue that we overestimate the requirements
for stabilizing North Korea because North Korean security forces could
perform many of the missions discussed above. True, many missions
could eventually be performed by local personnel. During the critical initial
days after a collapse, however, such forces should not be relied on. As in Iraq,
North Korea’s police and military could disappear. And even if they were
available, it would be risky to use them for the missions we describe. Many
might continue to engage in corruption and smuggling. North Korean personnel
would also need training and speciªc orders about the rules of engagement
under which they would be expected to operate. Furthermore, there is uncertainty
about whether the North Korean public would accept the authority of
these individuals, or whether they would seek revenge for their complicity in
the crimes of the Kim regime. For all of these reasons, the critical missions to
The Collapse of North Korea 111
77. “Ex-South Korean Presidents Convicted,” Time, August 26, 1996. Death sentences were eventually
commuted in these cases. Later, another former president, Roh Moo-hyun, accused of accepting
bribes during his presidency, apologized and committed suicide as investigations began.
stabilize North Korea should be performed by reliable personnel; local security
services should be employed only after careful planning, screening, and
training.
Composition of a Stabilization Force
Which countries might contribute forces to stabilizing a collapsed North
Korea? Although the event could threaten countries far ºung from the
peninsula—particularly if North Korea’s nuclear weapons go missing—
the countries most inclined to contribute personnel would be those involved
by virtue of geography and immediate strategic interest. First among these is
South Korea, which would likely seek uniªcation with the former North
Korean state. We therefore expect that most of the stabilization force would be
staffed by South Koreans. Furthermore, a stability operation is more likely to
succeed if its personnel understand the country, culture, history, and language.
78 Although it is uncertain how much legitimacy North Koreans would
accord South Korean forces, it would likely be more than that given to other
foreign forces. The bottom line is that North Korea’s collapse is a Korean problem,
which Koreans must take the lead in solving.
The United States is another highly interested actor that could become involved,
by virtue of its alliance ties, interests, and capabilities. It is a longtime
ally of South Korea, and its forces have trained for six decades to ªght military
operations alongside South Koreans on the peninsula. Beyond this, the United
States is a regional hegemon with a declared commitment to maintaining stability
in the Asia-Paciªc region. Advocates of continuing a U.S. strategy of primacy
or hegemony would argue that a close relationship with a uniªed Korea
would help to subvert Chinese inºuence on the peninsula, and that active U.S.
participation during a Korean transition would further this goal.79
Moreover, given its “global war on terror,” the United States could choose to
intervene out of particular concern for the “loose nukes” problem. The
U.S. military has unique WMD elimination units trained and equipped for
locating nuclear weapons within facilities and engaging in “sensitive site
exploitation”—the collection of intelligence about the nature and quantity of
WMD at a particular site. American intervention is already suggested by U.S.
International Security 36:2 112
78. On ethnicity and success in counterinsurgency operations, see Lyall, “Are Coethnics More Effective
Counterinsurgents?” Communication problems will still arise among Koreans. See Choe
Sang-hun, “Koreas: Divided by a Common Language,” International Herald Tribune, August 30,
2006.
79. On the potential for Korea’s strategic drift toward China, see David C. Kang, China Rising:
Peace, Power, and Order in East Asia (New York: Columbia University Press, 2007).
planning with South Korea, and by the conduct of joint exercises aimed at preparing
for emergencies associated with North Korean collapse, in particular,
locating, securing, and eliminating North Korea WMD.80
American participation is not assured, nor is it likely to be extensive. Hundreds
of thousands of U.S. troops are tied up in or rotating to Afghanistan and
Iraq, and in 2011 President Barack Obama’s administration initiated another
military commitment in Libya. In 2008 Secretary of Defense Robert Gates commented,
“The way that things have evolved in Korea, if there ever should be a
conºict, the main American contribution is not ground forces.”81 Furthermore,
the United States is facing a massive debt burden, and its people may be wary
of embarking on another costly foreign adventure.
Substantial U.S. participation could also undermine the success of a stability
operation in North Korea. Deep-seated anti-American sentiment exists
not only in North Korea—whose people have been raised on a diet of anti-
American vitriol—but also among many people in the South. Many Koreans
in both North and South blame their division on the United States.82 Coming
on the heels of the U.S. invasions of Afghanistan and Iraq, the image of
American troops swarming all over North Korea may create the perception
among some Koreans that the United States is essentially absorbing Korea into
its “empire” with Seoul’s blessing. A large U.S. troop contingent may therefore
wrest legitimacy from the stability force and accord it instead to any insurgents.
As Daniel Byman writes, “Since the best cause for insurgents to harness
is nationalism, direct and open U.S. support can undercut the legitimacy of a
government.”83 Furthermore, large numbers of U.S. troops on China’s border
would deeply antagonize Beijing. China’s willingness to cooperate in a Korean
settlement may depend in large part on U.S. behavior during Korea’s transition.
U.S. restraint and multilateral cooperation would help to reassure the
Chinese that the United States was not attempting to dominate the peninsula,
which may mollify Beijing’s attitude toward a future strategic relationship between
the United States and uniªed Korea.84
The Collapse of North Korea 113
80. Kwang-Tae Kim, “U.S. Anti-WMD Troops Join Military Drills in South Korea,” Associated
Press, March 11, 2010. On similar U.S. worries about, and contingency planning for, a loose nukes
problem in Pakistan, see Joby Warrick, “Pakistan Nuclear Security Questioned,” Washington Post,
November 11, 2007.
81. “U.S. Concerned about Possible Massive Outºow of N. Korean Refugees: Gates,” Yonhap
News Agency, September 25, 2008.
82. Selig S. Harrison, Korean Endgame: A Strategy for Reuniªcation and U.S. Disengagement (Princeton,
N.J.: Princeton University Press, 2000), p. 11; and B.R. Myers, The Cleanest Race: How North Koreans
See Themselves and Why It Matters (New York: Melville House, 2010).
83. Byman, “Understanding Proto-Insurgencies,” p. 196.
84. WikiLeaks cables suggested that “a new, younger generation of Chinese leaders ‘would be
comfortable with a reunited Korea controlled by Seoul and anchored to the United States in a beChina
is another highly interested actor by virtue of history and geography.
It fought in the Korean War on the North’s side, suffering hundreds of thousands
of casualties. Since then, Beijing has invested a great deal in the relationship,
funneling substantial quantities of food and energy aid to Pyongyang.
Chinese ªrms have also acquired substantial rights to minerals, ports, and
other resources in North Korea. China is North Korea’s sole military ally,
as speciªed in their 1961 Treaty of Friendship, Cooperation, and Military
Assistance.85
The Chinese dread a North Korean collapse for a variety of reasons. Because
of their shared border, China views the prospect of North Korean civil war or
loose nukes as a serious security threat.86 Chinese leaders worry about a massive
refugee ºow that would add to the tens of thousands of North Korean
refugees who already reside in China. Beijing is concerned about potential instability—
possibly separatism—in Jilin Province, which is in great part ethnically
Korean. Therefore, to stop a ºow of North Korean refugees, China may
send the PLA to its border, or even beyond the border into North Korean territory.
87 Beijing has already assigned PLA forces (instead of People’s Armed
Police, which guard other border areas) to guard the border region near North
Korea.88
Beijing also worries about North Korea’s collapse from a geopolitical standpoint.
The North Koreans “keep at bay the tens of thousands of U.S. troops stationed
in South Korea,” which reduces the U.S. ability to defend Taiwan.89
Beijing also sees North Korea as an important buffer zone between the Chinese
and U.S. spheres of inºuence. As scholar Shen Dingli argues, “If North Korea
were defeated, the eventual outcome could lead to Japan, South Korea, North
Korea, and Taiwan (a part of China) all aligning with the United States.”90
Beijing would thus be very interested in the orientation of a uniªed Korea: it
would want a seat at the table when Koreans discuss uniªcation, the prospect
International Security 36:2 114
nign alliance.’” David E. Sanger, “North Korea Keeps the World Guessing,” New York Times, November
29, 2010.
85. On Chinese business activities in North Korea, see Barbara Demick, “China Launches Economic
Projects in North Korea,” Los Angeles Times, June 10, 2011; and Lou Kilzer, “China Gains
Foothold via North Korea Port,” Pittsburgh Tribune-Review, February 20, 2011. On the Sino-North
Korean strategic relationship, see Harrison, Korean Endgame, pp. 311–312.
86. See Bonnie S. Glaser and Scott Snyder, with See-Won Byun and David J. Szerlip, “Responding
to Change on the Korean Peninsula: Impediments to U.S.-South Korea-China Coordination”
(Washington, D.C.: Center for Strategic and International Studies, May 2010), pp. 13–19.
87. Glaser, Snyder, and Park, “Keeping an Eye on an Unruly Neighbor”; and Stares and Wit, “Preparing
for Sudden Change in North Korea,” p. 7.
88. Thompson, “Border Burdens,” p. 16; and Shen Dingli, “North Korea’s Strategic Signiªcance to
China,” China Security, Autumn 2006, p. 47.
89. Ibid., p. 20.
90. Ibid., pp. 21–22.
of a U.S. alliance with a uniªed Korea, and the possibility of a continued U.S.
military presence there.91 For all of these reasons, China would be likely to join
a stabilization effort—or to act unilaterally if none were organized.
At the same time, China also has an interest in reassuring its neighbors of its
benign intentions at this important moment in its national growth. The need
for reassurance is particularly strong given that recent Chinese assertiveness
has undermined China’s earlier successful effort to reassure its neighbors of its
conciliation and “peaceful rise.”92 Distrust of China is rife in other countries—
some analysts warn of the precedent of Tibet and argue that China would be
interested in “establishing a puppet state” or in “fully incorporating North
Korea into China proper as a new Korean autonomous area.”93 South Koreans
worry about China’s so-called Northeast Project: a research project that claims
that ancient Korean kingdoms were part of China. Some analysts argue that
this effort represents an early stage of a Chinese land grab.94 A Korean transition
to uniªcation would give China an opportunity to demonstrate to its
neighbors and a carefully watching world that it does not seek territorial
expansion.
Japan shares many of the same interests as its neighbors: it too dreads
the prospect of loose nukes, and fears that a North Korean civil war would
create regional instability. Many North Koreans emigrated from Japan and still
have relatives there; in the event of a government collapse, some North
Koreans might seek refuge in Japan. The refugee crisis facing Japan, however,
would most likely be lessened by the geographical obstacles (generally more
onerous relative to ºight to China or South Korea) and by the anti-Japanese
sentiment long drilled into North Korean minds.
Signiªcant Japanese participation in stabilizing North Korea is unlikely and
probably undesirable. The Japanese people are likely to oppose direct participation;
the dispatch of even peacekeeping forces overseas remains controver-
The Collapse of North Korea 115
91. Harrison, Korean Endgame, pp. 324–327; and Andrei Lankov, “Why Beijing Props Up Pyongyang,”
New York Times, June 11, 2009.
92. On disputes about the South China Sea and the Senkaku/Diaoyu Islands, see Edward Wong,
“Chinese Navy Reaches Far, Unsettling the Region,” New York Times, June 14, 2011; and David E.
Sanger, “Three Faces of the New China,” New York Times, September 26, 2010. On Chinese fury at
the Nobel Prize award to dissident Liu Xiaobo, see Andrew Jacobs, “China, Angered by Peace
Prize, Blocks Celebration,” New York Times, October 9, 2010. On China’s “peaceful rise,” see Zheng
Bijian, “China’s ‘Peaceful Rise’ to Great Power Status,” Foreign Affairs Vol. 84, No. 5 (September/
October 2005), pp. 18–24; and Avery Goldstein, Rising to the Challenge: China’s Grand Strategy and
International Security (Stanford, Calif.: Stanford University Press, 2005), chap. 6.
93. Capt. Jonathan Stafford, “Finding America’s Role in a Collapsed North Korean State,” Military
Review, Vol. 88, No. 1 (January/February 2008), p. 99.
94. For discussion, see Gries, “The Koguryo Controversy, National Identity, and Sino-Korean Relations
Today”; “Koguryo or Goguryeo?” Korea Herald, September 1, 2004; and James Brooke,
“Seeking Peace in a Once and Future Kingdom,” New York Times, August 25, 2004.
sial.95 Japan’s military participation is also ill-advised because of regional
sensitivities about its past military aggression. In particular, Japan’s neighbors
vividly recall previous occasions in which Japan sent its military forces to
the peninsula during times of unrest. Japan, led by Hideyoshi Toyotomi, invaded
Korea twice in the late sixteenth century, and in 1894 Japan sent troops
to intervene in Korea’s Tonghak peasant rebellion. This led to the ªrst Sino-
Japanese war and to Japan’s thirty-ªve-year colonization of Korea. During the
Korean War, South Koreans stridently opposed any Japanese military support
against the North; President Syngman Rhee called the Japanese “a greater
threat than the Communists.”96 Japan’s troubled record of coming to terms
with its past violence remains a problematic issue in its regional relations.97
China would share these apprehensions, particularly given its own memories
of the brutal war it fought with Japan after 1937 and the atrocities committed
by Japanese soldiers in China. The Chinese would thus also react with alarm to
a Japanese military presence on the peninsula.
Although it is unlikely to participate directly in a stabilization effort, Japan
could make important nonmilitary contributions. It could allow stability forces
to use its base network to transport soldiers and supplies to Korea from Japan.
It could donate aid, particularly food and medicine. It might send civilian
medical personnel, aid workers, and possibly even police ofªcers to participate
in the stability operation. In the longer term, Japan could offer development
assistance and aid. A Korean transition would represent an opportunity
for Japan: Japanese generosity at such a momentous time in Korean history
could help to repair the still fragile relations between the two countries.
Russia has a long economic and strategic relationship with North Korea,
which frayed in the early 1990s with the collapse of the Soviet Union. Once
North Korea’s major trading partner and sponsor, the Soviet Union ceased
providing it with aid and subsidized petroleum in the late 1980s, a move that
in large part triggered North Korea’s downward economic spiral. Russia continues
to supply some petroleum and food aid to North Korea, although it has
severely curtailed its exports since Pyongyang’s 2006 nuclear test.
Russia has only a sliver of border with North Korea, with very little transportation
across it. In comparison to China, Russia faces a much smaller threat
of North Korean refugees or related problems. But Russia does have some eco-
International Security 36:2 116
95. Thomas U. Berger, Cultures of Antimilitarism: National Security in Germany and Japan (Baltimore,
Md.: Johns Hopkins University Press, 1998).
96. Quoted in Jennifer Lind, Sorry States: Apologies in International Politics (Ithaca, N.Y.: Cornell
University Press, 2008), p. 41.
97. Ibid., chap. 2; and Alexis Dudden, Troubled Apologies among Japan, Korea, and the United States
(New York: Columbia University Press, 2008).
nomic and political interests in North Korea, and in the event of a North
Korean collapse, it would probably want to assert its role as a major regional
power. Like China, Russia views North Korea as an important buffer to greater
U.S. inºuence in continental Asia, so it would be concerned about the strategic
implications of uniªcation. Russia does have a roughly 75,000-man ground
force in the Far East Military District,98 and it could decide to commit part of it
to becoming involved in North Korea if China, South Korea, and the United
States were to intervene. Still, it would most likely be a minor player in such
an effort.
Finally, the United Nations could play a useful role in the stabilization of
North Korea. South Korea and others could seek a UN Security Council resolution
to authorize a stability force. David Edelstein argues that multilateralism
is a key attribute of successful military occupations, “because it makes an
occupier’s pledge to withdraw more credible. . . . Presumably, the international
community is not interested in forming an empire, so an occupied population
can be conªdent that a multilateral occupation will, in fact, come to an end in a
reasonable period of time.”99 In a post-collapse North Korea, a multinational
force with a credible pledge to withdraw would have several advantages. It
would reassure Koreans that they would not be subject to never-ending occupation,
assuage the Chinese that U.S. participation in a stabilization effort
would not lead to U.S. domination of the peninsula, and reduce U.S. fears that
China would try to leave military forces in northern Korea.
Conclusion
No one knows when the government in North Korea might collapse, and indeed
the country may limp along for another few decades. But even if one believes
that a North Korean government collapse is unlikely, the magnitude of
the problems that it might cause makes this contingency worth studying. Absent
careful joint preparation for such an event, a transitioning Korea could be
a chaotic and dangerous place in which the militaries of two nuclear-armed
great powers are jostled together.
To further the goal of increased discussion and joint planning, this article described
the instability that might accompany a North Korean government collapse,
outlined the missions that countries might choose to perform to mitigate
The Collapse of North Korea 117
98. International Institute for Strategic Studies (IISS), The Military Balance 2010 (London: IISS,
2010), p. 230.
99. David M. Edelstein, “Occupational Hazards: Why Military Occupations Succeed or Fail,” International
Security, Vol. 29, No. 1 (Summer 2004), p. 72.
that instability, and estimated the military requirements of those missions.
Based on fairly optimistic assumptions about how a collapse would occur, we
estimate that 260,000–400,000 troops would be necessary to staff the missions
described here.
This analysis leads to several implications. First, the possibility of a North
Korean collapse should inform South Korea’s contemporary defense planning.
Given the current size of its military, South Korea could ªeld the personnel
necessary to perform the missions described above, especially if its reserves
were mobilized. But in its Defense Reform Plan 2020, the government plans to
trim South Korea’s ground forces from roughly 550,000 now (Marines and
Army combined) to about 420,000 by 2020—too low a level given the requirements
estimated here.100 Furthermore, in light of this contingency, the government
should consider reforming South Korea’s military reserve structure.
Access to the reserves is currently limited to full mobilization, which would
withdraw a vast number of people and resources (e.g., trucks) from the civilian
economy. The government should establish a selective mobilization system
that could more ºexibly respond to developing circumstances.101 It should
therefore consider the number of forces it is prepared to ªeld and discuss with
other countries how much assistance might be available (and desirable). Then
the government should reform South Korea’s ground forces and reserve structure
accordingly.
Second, South Korea and others might also begin reaching out to North
Korean military and governmental ofªcials earlier than the United States did
in Iraq before the 2003 invasion. For example, the U.S. government tried to
reach Iraqi ªeld commanders weeks before Operation Iraqi Freedom, promising
to take care of them and their families if they did not oppose the impending
invasion. That effort, however, came as too little too late and did not
give the Iraqis time to establish its credibility and recognize its potential
advantages.
Third, this analysis underscores the need for multilateral planning—not
only between South Korea and the United States, but also with China.
American and South Korean ofªcials should understand that China will be
likely to intervene if a North Korean government collapse fuels instability on
the Chinese border. Given the potential magnitude of operations to stabilize
North Korea, South Korean and U.S. leaders may welcome some Chinese participation.
For example, Chinese forces might logically assume border control
International Security 36:2 118
100. Bennett, “A Brief Analysis of the Republic of Korea’s Defense Reform Plan.”
101. BruceW. Bennett, “Future ROK Army Development: NE Asia Security in 2020,” unpublished
paper, RAND, Santa Monica, California, 2008.
responsibilities on the Sino-North Korean border (a task that we estimate
would require 24,000 troops). China’s commendable management of the 2009
refugee ºow from Myanmar suggests that China has prepared itself well to
undertake this mission.102 Beijing could contribute forces to a multilateral stability
operation, and it could assist in WMD elimination (especially given that
many sites are much closer to China than to South Korea).
Conversely, if American and South Korean leaders oppose Chinese involvement,
they should be prepared to conduct the missions to stabilize North
Korea themselves—and should communicate their intentions and goals to
Beijing. While ultimately South Korea and the United States may have no control
over a Chinese decision to intervene, advance and joint planning can inºuence
the likelihood of Chinese intervention as well as the potential for
dangerous escalation.
American and South Korean diplomats already understand the merits of advance
planning with China; they have struggled for years to initiate such discussions
with a recalcitrant Beijing.103 Concerned governments, however,
could support discussions at the track II level—for example, between scholars
and think-tank analysts, and among retired military ofªcials. In addition,
ofªcials should not abandon their efforts to coordinate with Beijing at the
ofªcial level, which would do the most to reduce the risk of misperception and
escalation.
The Collapse of North Korea 119
102. Thompson, “Border Burdens.”
103. Stares and Wit, “Preparing for Sudden Change in North Korea,” p. 8.